No Choice: "Weapons-effect" paralysis

By Dave Kopel & Paul Gallant & Joanne Eisen

National Review Online. April 17, 2002 9:55 a.m. More by Kopel on psychology and guns.

On September 16th, Reuters revealed that in the midst of the Sept. 11 hijackings, President Bush had ordered U.S. fighter pilots to shoot down any airliner acting in a suspicious manner. Since then, a fleet of 250 military planes, including AWACS surveillance aircraft and refueling tankers, has flown nearly 20,000 missions over U.S. cities. To date, military aircraft have responded to more than 350 "air events" — instances when a plane flew off course or failed to maintain radio contact.

Under current guidelines, the order to shoot down an airliner considered to be a security threat would come from President Bush, but, if time were critical, several generals in the chain of command have the authority to issue a shoot-down order.

On March 18, Defense Department officials announced that only the air space above the nation's capitol would be subject to round-the-clock patrols. New York senators Schumer and Clinton protested the decision. Two days later, Homeland Security Director Tom Ridge characterized the earlier reports of curtailed surveillance as "inaccurate," and gave assurances that some form of military air patrols would continue over New York. Governor Pataki hailed this latest announcement "good news."

Early discussions after Sept. 11 about how best to deal with terrorists in the air focused on whether to allow armed pilots in the cockpit. There was hardly a thought about allowing off-duty cops, and civilians with concealed-carry licenses, as armed passengers aboard airliners. But the idea of armed pilots now appears officially off the table, with Transportation Secretary Norman Mineta's statement on March 4, "I don't feel we should have lethal weapons in the cockpit."

What has been utterly incomprehensible is that the federal government considers destroying a hijacked airliner, and thereby killing every single passenger, more palatable than allowing the arming of pilots or passengers.

Why does the slim possibility of a bad outcome from firearms onboard a plane (e.g. an innocent passenger shot by mistake) provoke so much more fear than does the death of all those aboard the plane? What about the likelihood of additional ground casualties resulting from a plane being blasted out of the skies?

Twenty-two years ago, Dr. Lester Adelson wrote in The Pharos, a medical-society journal: "The accessibility of a firearm permits the instantaneous metamorphosis of a law-abiding person into a murderer." Adelson had perfectly articulated what has become known as the "weapons effect" hypothesis.

Whether called the "weapons effect," the "instrumentality theory," or the "accessibility thesis," the premise is the same: Guns provoke impulsive, violent responses, and the presence of firearms anywhere (except in the hands of government employees) is to be feared. The illogical fear of firearms on airplanes is evidence that the weapons-effect hypothesis has thoroughly permeated our culture.

In 1967, the founding article on the weapons-effect hypothesis was published in the Journal of Personality and Social Psychology. Authored by psychologists Leonard Berkowitz and Anthony LePage, "Weapons as Aggression-Eliciting Stimuli" summarized the results of their laboratory experiment on 100 male undergraduate students at the University of Wisconsin. Berkowitz and LePage hypothesized that the mere sight of a firearm could trigger aggression from an "already angered" person because of the learned association between violence and guns.

In their experiment, each subject was paired off with a partner. Subjects were not informed that their partners were actually confederates of the researchers.

During the experiment, subject and confederate were placed in separate rooms. Mild electric shocks (the "stress" component), administered between subject and confederate, were supposed to be based on performance of a task; the greater the number of shocks administered, the poorer job performance was (supposedly) judged to be. However, the number of shocks administered by the confederate was predetermined — and independent of the actual performance of the subject.

This way, Berkowitz and LePage reasoned, two groups of subjects would be created: "angered" subjects, each receiving a larger number of shocks thereby making them "physically uncomfortable" (and, the researchers believed, made to feel "humiliated"); and "unangered" subjects, who received just one shock from their (confederate) partner.

For the second part of the experiment, the pair exchanged rooms, and the subject was placed in the room containing the "shock key." Now it was the subject's turn to grade the (confederate) partner's job performance. Some of the time, a firearm was casually (and, the subject was told, inadvertently) left in plain view nearby.

Berkowitz and LePage translated the number of shocks administered to the confederate lab partner as a measure of aggressive behavior on the part of the subject. What they found was that the "angered" group of subjects administered a greater number of electric shocks to the confederate in return (and held the "shock key" down longer), when either a shotgun or a revolver was deliberately left lying in plain view in the room next to the subject, compared to when nothing or neutral objects (badminton racquets and shuttlecocks) were left.

Berkowitz and LePage concluded from their findings that "the presence of the weapon might have elicited an intense aggressive reaction from the person with the gun…" A year later, in discussing his experiment in Psychology Today (1968), Berkowitz flatly stated: "Guns not only permit violence, they can stimulate it as well. The finger pulls the trigger, but the trigger may also be pulling the finger" ("Impulse, Aggression and the Gun").

However as Arthur Kellerman, a prominent researcher, has pointed out: "the strongest proof of the validity of any study is independent replication by others." ("Response to Kleck," Homicide Studies, p.276, Aug 2001). Subsequent attempts to reproduce the findings of Berkowitz and LePage met mixed results. Criminologists Hans Toch and Alan Lizotte noted that "many studies have failed to replicate the Berkowitz and LePage findings, and some have even reported opposite findings."

A number of weapons-effect researchers later reported that an "artifact" had surfaced in the course of their experiments. Interestingly, many of the subjects saw the weapon and instinctively understood that there was an unexplained motive for its presence in the experiment, despite how clever or elaborate a cover story was provided by the researchers.

Dolf Zillmann (Hostility and Aggression, 1979), in discussing this problem, pointed out that there was no discernible pattern to this source of contamination, and it was unclear why some subjects "were noncooperative, seeming eager to thwart the experimenter's efforts and to 'louse up' the experiment." One pair of researchers even noted that hostility was directed toward them because of the fabricated cover story, rather than toward the lab partner.

Criminologist Gary Kleck reviewed twenty-one "weapons effect" published studies, and concluded that "none of the studies provided any evidence directly supporting the idea that possessing a gun encourages physical aggression, or that the 'trigger pulls the finger.'" As Kleck further observed, "the more closely the experiments simulated real-world situations…the less likely they were to support the weapons hypothesis." That shouldn't come as any surprise. Berkowitz and LePage carried out their experiment in a laboratory setting, and the consequences of the actions of the experimental aggressors were neither serious nor permanent. It is quite another thing when the consequences of one's actions can be lethal, and when there is a significant risk of punishment by the law.

Another means of determining the validity of a hypothesis is to examine how well it predicts the future, compared to what would be expected on the basis of chance alone. In the case of the weapons effect hypothesis, its real-world predictions fail abysmally.

If guns facilitated the transformation of ordinary people into killers, it would be reasonable to expect to find ordinary people killing victims all over the country. They are not. Instead, we know that the best predictor of violent behavior by a person is not proximity to a weapon, but prior violent behavior. As criminologists Don Kates and Dan Polsby have detailed, "perpetrators of homicide are anything but ordinary people...It is...an extravagant falsification of reality to claim, as some researchers have, that such individuals 'would be considered law-abiding citizens prior to their pulling the trigger.'"

Kleck points out that

very few homicides are committed by people who have no prior history of violence. The image of the model citizen who one day goes berserk and kills a family member is largely a media-created myth maintained by newspeople enamored with the dramatic contrast between extremely violent acts and supposedly peaceful backgrounds…The apparently "nonviolent" killer is a rare exception to a rather mundane general rule: People who are seriously violent in the present almost invariably have been seriously violent in the past.

It would also be reasonable to expect that, if the weapons-effect hypothesis were correct, as the number of guns in America rises, so should firearm-related violence. However, the last quarter-century, the per-capita supply of firearms has doubled, while homicide and other gun crimes have declined.

During the 1990s, many states eased restrictions on the concealed-carry of handguns; this same time period was characterized, not by an increase in violent crime, but by a dramatic decrease. On November 18, 1999, the Centers for Disease Control and Prevention reported that, between 1993 and 1997, the firearm-related death rate dropped to the lowest level in more than 30 years. And on August 27, 2000, the U.S. Department of Justice reported that the violent crime rate had fallen to the lowest rate since 1973.

In a landmark 1995 study on defensive gun use in America, Gary Kleck and Marc Gertz found that only 24 percent of people who use guns defensively actually fired the gun. As Kleck later commented (Targeting Guns: Firearms and Their Control, 1997), "More commonly, guns are merely pointed at another person, or perhaps only referred to ('I've got a gun') or displayed, and this is sufficient to accomplish the ends of the user.…"

How, then, is it possible to reconcile the weapons-effect hypothesis with the finding that even when being criminally attacked or threatened, 76 percent of defensive gun user did not fire his gun?

Weapons-effect proponents use their model as a justification for decreasing firearm availability to everyone, since any one of us might become an unpredictable perpetrator of firearm rage. Kleck calls this the "blunderbuss" approach premised on the supposition that "it is impossible to distinguish between low-risk and high-risk candidates for gun ownership, that everyone is a potential killer, and that serious acts of violence and other criminal acts committed with guns are common among people with no previous record of violence."

Such a rationale forms the basis of programs such as gun-surrender programs, encouraging law-abiding gun owners to turn in their guns to the government, in exchange for money or some other reward. Such programs make sense if the weapons-effect hypothesis is valid. If so, lowered firearm density would lower levels of firearm-related violence, because even if criminals did not give up their guns, fewer non-criminals would succumb to the "aggression-evoking" madness caused by proximity to a firearm.

But has that been the case?

In April 2000, the Clinton administration allocated $2.6 million to fund the BuyBack America campaign, an 84-community program designed to "buy back unwanted guns and raise awareness about gun safety." (The program was not really a "buyback", because the government had not originally owned the guns.) Kansas City, Kansas, mayor Carol Marinovich promised: "The gun buyback program is an important step toward making our community safer."

However, such claims are invariably made without any social-science evidence. Gun-control scholar Dr. Garen Wintemute offers one rationale as to why research on gun-surrender programs does not show a reduction in violence: "Buybacks remove generally no more than 1 or 2 percent of the guns estimated to be in the community." It is because of such insignificant numbers that "There has never been any effect on crime results seen."

However, the result of removing a large quantity of firearms from a population has been studied. In the March 1999 British Medical Journal (BMJ), authors David Meddings and Stephanie O'Connor measured the incidence of weapon injuries before and after a U.N.-mediated peace agreement in northwestern Cambodia in the early 1990s. Although it was estimated that "around 25-50% of [Cambodia's combatant factions were]...believed to have been disarmed" during the peacekeeping operation, and although a stable government was left in place at the time of departure of the U.N., no reduction in firearm-related violence was observed. What the BMJ authors found was that firearm-related injuries rose.

If the weapons-effect hypothesis were valid, and if 25-50 percent of the available firearms were taken out of circulation, wouldn't we have seen some decrease in firearm-related violence?

The weapons effect, like 19th-century theories of the racial inferiority of nonwhites, is a form of social prejudice masquerading as science. Weapons-effect proponents would have us believe that ordinary American gun owners are like Pavlov's dogs learning to salivate upon hearing a bell; put them near a gun, and they will end up shooting themselves or some other innocent person. Leonard Berkowitz put it this way: "Gun control may not be too effective in protecting ordinary citizens against criminals or Presidents against assassins, but it may, nevertheless, save some ordinary citizens from other ordinary citizens like themselves" ("How Guns Control Us"; Psychology Today, June 1981). Such a profoundly pessimistic view of human nature presumes that most people are incapable of controlling their actions.

The theory also amounts to an excuse for murderers, who are allowed to claim, "It wasn't my fault; the gun made me do it." Thus, the weapons-effect hypothesis is one variation of the modern tendency to manufacture excuses for criminals: muggers can't be held morally accountable because they grew up in an environment of poverty; Andrea Yates isn't responsible for her actions because she was driven insane by being a stay-at-home mother with a husband who had a job; the Columbine murderers were just the products of Denver's militaristic culture (according to reports of Michael Moore's forthcoming movie Bowling for Columbine). Does the defeatist attitude of environmental determinism actually promote crime and other evil, by telling people that they can't be expected to be responsible for their conduct?

Albert Schweitzer had it right: "Man must cease attributing his problems to his environment, and learn again to exercise his will — his personal responsibility in the realm of faith and morals."